Kubernetes cluster persistent storage
When creating LKE clusters you add nodes as follows:
| Plan | CPUs | Storage | Ram |
|---|---|---|---|
| Linode 2GB | 1 | 50GB | 2GB |
| Linode 4GB | 2 | 80GB | 4GB |
| … | … | … | … |
| Linode 192GB | 32 | 3840GB | 192GB |
Given that containers are 'mortal' why do these plans include increasing storage? I was under the impression that the storage on the node is used to contain containers which should not be that big and the rest of the data would be mounted into the container from block storage. How can the extra storage be used?
2 Replies
I believe I'm right in saying that the Kubernetes nodes are essentially just Linodes with a preloaded image.
You can treat these nodes just the same as any other Linode - such as SSH'ing into them, and installing additional software, with the benefit of them being pre-configured to your LKE cluster.
For me it does make sense that they match the same plans as "non-LKE" Linodes; you don't have the confusion of why you'd get lower allocated storage space than if you spun up the same spec Linode outside of LKE.
I was under the impression that the storage on the node is used to contain containers which should not be that big and the rest of the data would be mounted into the container from block storage. How can the extra storage be used?
AFAIK there is no requirement to use block storage with LKE - you could use the rest of the Linode's SSD and mount permanent storage into a container I presume? Although granted that container wouldn't be functional outside of the allocate node I guess?
